Pathogenesis of a subset of craniofacial anomalies (frontonasal dysplasia, microphthalmia) and its likely molecular basis are the subject of this proposal. Vitamin A deficiency is endemic in some parts of the world and may contribute to a proportion of the large majority of birth defects worldwide which have currently no known cause. Evidence for the involvement of vitamin A in congenital craniofacial anomalies comes from animal models based on nutritional deprivation or gene "knockout" mice which are missing one or more of the nuclear retinoid receptors (retinoic acid receptor-- RARs, and retinoid X receptors--RXRs). RARs and RXRs are ligand activated transcription factors. All - trans- and 9-cis retinoic acid (RA) are endogenous metabolites of vitamin A which serve as ligands for the RARs and RXRs. To elucidate molecular pathogenesis of developmental anomalies and to dissociate RA-dependent from RAR-independent events, we propose a new animal model more suitable for this purpose than the other approaches. In this model a synthetic retinoid antagonist, which binds one or all three RARs (alpha, beta, gamma) with high efficiency but fails to transactivate through them, is fed to pregnant mice to render the respective vitamin A signaling pathway(s) inoperable at discrete embryonic stages. Initial studies have yielded fetuses with versions of frontonasal dysplasia and small eyes (or no eyes) after treatment with a universal RAR antagonist on 8 day post coitum (dpc). The aims are: First, to characterize the anomalies in embryos exposed to the universal antagonist at different stages of organogenesis, and to ascertain any changes in phenotype ascribable to individual receptor subtype inactivation by more specific antagonists. Second, to address the hypothesis that the antagonist "starves" the mice of vitamin A-dependent functions. This will be done by assessing the reversibility of antagonist actions by RAR agonists, by quantification of the levels of antagonist in sensitive tissues in presence or absence of the RAR agonists, by quantification of the levels of the antagonists in the sensitive tissues in presence or absence of the RAR agonists, and by assaying the levels of several mRNAs (e.g., CRBP-1, RAR beta2) known to be modulated in vitamin A deficient animals. Third, to characterize the cell/tissue site(s) of disruption initiated in 8 dpc embryos by the RAR antagonists and establish any deviations in gene expression patterns as likely contributors to malformations of the orofacial and cranial regions. It is hoped that fresh information will be elicited from this new animal model.